bug-autoconf
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: autoconf 2.65: AC_PROG_F77 code executed though in a test that is al


From: Ralf Wildenhues
Subject: Re: autoconf 2.65: AC_PROG_F77 code executed though in a test that is always false
Date: Sat, 6 Feb 2010 14:45:33 +0100
User-agent: Mutt/1.5.20 (2009-12-10)

Hello Vincent,

thanks for the report.

* Vincent Lefevre wrote on Fri, Feb 05, 2010 at 02:51:25PM CET:
> I can reproduce it under Mac OS X (with autoconf 2.65).
> 
> Note that this is a regression. On a Debian/stable (lenny) machine
> with autoconf 2.61-8, there is no such problem.
[...]
> On 2010-02-05 14:14:56 +0100, Vincent Lefevre wrote:
> > In MPFR, we had in configure.in:
> > 
> > if test -n "$variable_indefinie" ; then
> >  AC_PROG_CXX([g++])
> >  AC_PROG_F77([g77])
> >  AC_PROG_CXXCPP
> > fi
> > 
> > This doesn't seem to be forbidden according to the autoconf
> > documentation. Even though the test
> > 
> >   test -n "$variable_indefinie"
> > 
> > is always false, the configure script checks for the GNU Fortran 77
> > compiler.

Confirmed with Libtool 2.2.6b.  This issue is a bug in Libtool, and
fixed in git Libtool; the fix is this commit:
<http://git.savannah.gnu.org/cgit/libtool.git/commit/?id=57013d8e689ed32f0a4fb8d3b0549ed5e793e2d6>

Cheers,
Ralf




reply via email to

[Prev in Thread] Current Thread [Next in Thread]